Project Summary The goal of the proposed project is to investigate the cellular and molecular consequences of splicing factor (SF) mutations in myelodysplastic syndromes (MDS). These blood diseases are characterized by ineffective hematopoiesis manifested as peripheral blood cytopenias and dysplastic morphological changes and increased propensity for progression to acute leukemia. Recently, somatic mutations in RNA binding proteins with known roles in splicing, commonly referred to as splicing factors (with the three most common being SF3B1 K700E, U2AF1 S34F and SRSF2 P95L), were revealed as the most common class of mutations in MDS affecting more than half of patients. SF mutations occur early in the disease and are, therefore, likely critical for its pathogenesis and attractive therapeutic targets. Early studies using primary patient cells, mouse models and ectopic expression in cell lines using RNA-seq suggest that the mutations alter the RNA binding specificity of the mutant SFs. However, the downstream events that are critical for the disease pathogenesis are still elusive and the RNA binding specificity of the mutant factors has yet to be directly tested. Population genetics studies have shown that different mutations co-occurring with SF mutations are associated with different clinical outcomes (poor or favorable prognosis). Co-occurrence of DNMT3A with SF3B1 mutations is associated with good prognosis, whereas co-mutations of SF3B1 with either RUNX1 or ASXL1 confer poor prognosis. However, the mechanisms underlying these genetic interactions are not understood. The Papapetrou lab has pioneered the modeling of MDS using induced pluripotent stem cell (iPSC) models. Human iPSC models are particularly attractive for the investigation of SF mutations, as alternative splicing isoforms show poor conservation between mice and humans. I propose to develop novel iPSC-based models of SF mutations and cooperating mutations to study the role of these mutations and their genetic interactions in MDS pathogenesis. Specifically, I propose to: (1) generate iPSC lines with the three most common canonical SF mutations SRSF2 P95L, SF3B1 K700E and U2AF1 S34F using CRISPR/Cas9 and characterize them phenotypically and molecularly using RNA-seq and eCLIP- seq) and (2) generate SF3B1 K700E iPSC lines with additional mutations in either DNMT3A, RUNX1, or ASXL1 and characterize their hematopoietic phenotypes and transcriptomes. This work can generate new hypotheses about the mechanisms by which SF mutations drive MDS and reveal principles of mutational cooperativity in MDS pathogenesis.